Extended Longevity of the Pre-emerged Adult Cat Flea (Siphonaptera: Pulicidae) and Factors Stimulating Emergence from the Pupal Cocoon J. SILVERMAN 1 AND M. K. RUST Department of Entomology, University of California, Riverside, California 92521-0314 Ann. Entomol. Soc. Am. 78: 763-768 (1985) ABSTRACT After the pupal-imaginal molt, the pre-emerged adult cat flea, Ctenocephalides felis felis (Bouche), remains quiescent inside the cocoon for varying periods of time. It has a lower respiratory demand than the emerged adult and its survival is considerably longer. Pressure and heat stimulate rapid emergence from the cocoon. In the absence of these stimuli adults emerge gradually over several weeks, depending on ambient temperature, with the length of time spent in the cocoon related to prepupal weight. This pharate stage facilitates flea-host contact by maximizing the active life span of the adult flea and minimizing nonhost-induced emergence. ADULT CAT FLEAS, Ctenocephalides felis felis (Bouche), like other fleas and hematophagous arthropods, must locate a suitable vertebrate host to complete its life cycle. Often the definitive hosts of C. /. felis (Hopkins and Rothschild 1953), such as feral cats, mustelids, and dogs, do not always return in a regular or predictable manner to their nests or resting areas where immature fleas develop. Furthermore, emerged unfed adult fleas live only ca. 1 week except at high humidities and cool temperatures (Silverman et al. 1981). The ephemeral presence of the host, the necessity of a blood meal, and the relatively short adult longevity require an adaptable lifestyle that takes these factors into account. Several researchers, including Bacot (1914), Karandikar and Munshi (1950), and Silverman et al. (1981), have observed adult fleas remaining quiescent for prolonged periods within the pupal cocoon before emergence. Herein we report on several characteristics of this unusual life stage, as well as factors that induce adult emergence. The adaptive significance of this stage, henceforth referred to as the pre-emerged adult, relative to the survival of a short-lived adult ectoparasite is discussed. Materials and Methods Experiment 1. Effects of the Cocoon on Adult Longevity and Emergence. Fleas used in these experiments were obtained from a stock culture reared as described by Silverman et al. (1981). To determine adult flea survival, pre-emerged and emerged adults were held at ca. 2 (anhydrous CaSOJ and 100% RH. Adult fleas were obtained 1 Present address: American Cyanamid Company, Clifton, NJ 07015. from third instars that had spun cocoons in either individual gelatin capsules for 2% RH tests or 2-cm 3 glass vials for 100% RH tests (gelatin melts at high RH). Before attaining the adult stage (pre-emerged or emerged), immature fleas were held inside chambers maintained at 17 C and 75% RH. Because cocoons were normally constructed against the wall of the transparent gelatin capsule or glass vial, the development of fleas could be observed. When >80% of the fleas developed into preemerged adults (ca. 8 days after becoming third instars), 100 cocoons were dissected to obtain adults. Fifty emerged adults and 50 cocoons containing pre-emerged adults were transferred to each relative humidity at 16 C. Mortality of emerged adults was recorded daily. Pre-emerged adult mortality in dry air was assessed by dissecting 10 cocoons on each of 5 successive weeks. Thefirst emerged adult maintained in saturated air died on day 42; thereafter, 10 randomly selected cocoons were dissected weekly to determine pre-emerged adult mortality. When adults began regularly emerging from their cocoons, fewer pre-emerged adults (three to seven) were dissected each week. Cocoon water-vapor permeability was determined by cutting a flap into the gelatin capsule wall where a cocoon had been spun. The pupa was removed and a small piece of humidity-sensitive cobalt thiocyanate paper was inserted. The flap was replaced and sealed with silicon cement. The cocoons (n = 30) were held at 10% RH for 1 h, then exposed to 75% RH for 1 h. The color of the paper was noted and compared with the RH-sensitive paper outside the cocoon. Experiment 2. A Comparison of Metabolic Activity Between Pre-emerged and Emerged Adults. Metabolic activity of pre-emerged adults was measured as CO 2 production. Newly emerged adults (20-30) were placed in each of six sealed 763
764 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 78, no. 6 6-ml glass hypovials. Three vials were wrapped with black tape to darken the inside completely, and held in ring-stand clamps to lessen stimulated adult activity induced by visual and physical stimuli. Three transparent vials with fleas were shaken every hour for 10 s to increase the activity of the adults. Twenty-five pre-emerged adults were sealed in each of three transparent hypovials. After 8 h at 22 ± 1 C, air samples (1 cm 3 ) taken from each vial with a gas-tight syringe were analyzed for CO 2) using a gas chromatograph (Aerograph Model A-110-C) equipped with a thermal conductivity detector and a silica gel column (0.64 cm diam by 1.52 m). The oven temperature was 125 C and the flow rate of helium over a 2150-mV filament was 68 ml/min. Peak height was correlated with amount of CO 2 expired. Only 10% of the previously unemerged adults emerged within the exposure period. The following formula corrects for CO 2 release due to emerged fleas: x ng CO 2 /unemerged adult = [/ng CO 2 /vial (no. emerged adults) (x fig CO 2 /undisturbed adults)]/no. unemerged adults. The function of the pupal cocoon as a physical barrier to adult emergence was determined by removing portions of cocoons and determining the subsequent rate of emergence. Cocoons containing ca. 6-day-old pupae were dissected to expose the entire flea, only the cephalic region, or only the pygidium. Some 26-39 pupae for each condition and 20 unaltered cocoons were held individually in gelatin capsules kept at 22 C and 45 ± 5% RH. Daily adult emergence was recorded. Experiment 3. Effects of Temperature and Humidity on Rates of Adult Emergence. To determine the rate of adult emergence at various constant temperatures and constant 75% RH, 100 pre-emerged adult fleas per temperature were held in individual gelatin capsules at 32, 27, 21, 16, or 11 C until emergence. Capsules were adhered to cardboard with double-stick adhesive tape and were examined weekly. The effect of various relative humidities at constant 27 C on adult emergence was determined by confining 100 preemerged adults per RH as above at 91, 75, or 33% RH. Humidities were maintained with the appropriate saturated salt solutions (Winston and Bates 1960). The number of emerged adults was counted every 3 days. Eggs and larvae were reared at different humidities to determine if fleas reared at lower humidities emerged sooner than fleas reared at high humidities. Several hundred eggs placed in 75-ml plastic cups provisioned with a mixture of 20 g of larval rearing media and sand were held at 27 C in each of three chambers maintained at 91, 71, and 52% RH. Third instars developing at each relative humidity were transferred to each of 100 individual gelatin capsules containing a small amount of sand. The capsules were adhered to cardboard and returned to each humidity chamber. To delay emergence and help magnify possible differences in the emergence rates at various relative humidities, each chamber was kept at 21 C after pupae darkened in the capsules. Lowering the temperature inside the chambers increased RH <2%. Adult emergence was recorded daily. Water content of prepupae held at each RH was determined gravimetrically. Ten prepupae removed from cocoons (five replications) were weighed immediately, dried at 100 C and 0% RH (anhydrous CaSOJ for 24 h, and reweighed. Experiment 4. Effect of Metabolic Reserves on Adult Emergence Times. The average and median number of days for emergence of adults developing from fully nourished larvae (food provided throughout larval development) were compared with those developing from larvae deprived of food for 3 days following eclosion from the egg. Most incompletely nourished larvae died. Consequently, only 50 were held for adult emergence versus 100 larvae that had not been deprived of food. Two larvae were placed in each capsule. Fleas were held at 21 C and 75% RH in gelatin capsules and adult emergence was recorded daily. The wet and dry weights of prepupae that developed from fully nourished or incompletely nourished larvae were determined. Since many starved larvae died in the preceding test, another experiment was designed to determine the effect of water expenditure and metabolic reserves on the average time for adult emergence. After 1-day-old cocoons were probed with a dissecting needle, the larvae abandoned their cocoons and spun new cocoons within 24 h. The wet and dry weights of prepupae developing from larvae that had spun one or two cocoons were determined as in the previous experiment. A comparison of emergence times for adults derived from larvae spinning one or two cocoons was determined by holding 100 pre-emerged adults (2 per gelatin capsule) at 21 C and 75% RH and counting adults each day. Experiment 5. Vibration as a Stimulus for Adult Emergence. The effect of vibrations on emergence from the cocoon was studied by subjecting pre-emerged adults to vibrations. Twenty gelatin capsules, each containing a pre-emerged adult, were fastened to the top of an inverted plastic food container with double-stick masking tape. Vibrations were produced by holding the top of a vibrating engraver against the container for 20 s. After 2 min the vibration sequence was repeated, and the number of emerged adults was counted 10 min later. Experiment 6. Pressure as a Stimulus for Adult Emergence. Pre-emerged adults emerged from a cocoon within 5 s if the cocoon was gently squeezed. A device applying weight against the cocoon was used to determine the effect of pressure on flea emergence. A rigid plastic tube (1 cm diam by 20 cm), closed at one end with a plastic screw-cap, was filled with up to 50 g sand. Additional weight was obtained by adding sand to a
November 1985 SlLVERMAN AND RUST: LONGEVITY IN CAT FLEA 765 funnel on top of the column. The weighted column, supported with a ring-stand clamp, was kept in vertical alignment inside a plastic tube held by another clamp. The column was lowered onto single cocoons containing adult fleas placed on a glass plate for 1 or 30 s. Each cocoon was observed 30 s for adult emergence after pressure was applied. Thirty cocoons were tested for each amount of pressure. The effect of various amounts of pressure on adult emergence was determined by applying 13, 26, or 38 g/cm 2 for 1 s up to five times at 1-min intervals. Ten cocoons were tested with each amount of pressure. Since cat fleas typically develop in carpets, incorporating fibers into the cocoons, adult flea emergence as a consequence of a human walking across carpet infested with fleas was also investigated. Fifty third-instar larvae were placed on each of five disks (9 cm diam) of nylon carpet (1.5 cm tall and 11 strands per cm 2 ) held in a chamber and allowed to pupate. A hole (9 cm diam) was cut in the center of a 1-m 2 piece of white carpet to accommodate the infested disks. At the beginning of the test, 12-23 pre-emerged adults per disk were counted. The carpet was then walked on, with the front Vz of the shoe (30.2 cm 2 ) contacting the disk. The force exerted on the carpet by a 75-kg man was ca. 2.5 kg per cm 2. Five successive amounts of pressure were administered at 1-min intervals for each of the five disks. Adults that emerged were easily spotted on the white background after each amount of pressure was applied and the number of emerged adults was counted. Experiment 7. Pressure and Warmth as Stimulus for Adult Emergence. All life stages of fleas are often abundant near the bedding of flea-inf ested pets. To simulate the influence of an animal resting on a pre-emerged adult, the combined effect of warm body temperatures and physical force on emergence was evaluated. Cocoons containing pre-emerged adult fleas were placed individually on the bottom of an inverted glass beaker warmed in a water bath and were observed for 30 s to determine rapid heat-induced emergence before applying pressure. The body temperature of cats and dogs is between 38 and 39 C. The lower surface temperatures were tested inasmuch as contact between host and cocoon is indirect because the substrate in which the cocoon is resting dissipates much of the body heat. Three replicates of 10 cocoons each were tested at each of the following surface temperatures: 38, 35, 32, 29, and 25 C. If there was no emergence, a 38 g/cm 2 force (determined from experiment no. 6) was administered for 1 s and the number of fleas subsequently emerging was counted for 30 s. Results All pre-emerged adults survived in cool dry air for >35 days, whereas 90% of emerged adults died 00 80 60 40 20 - ' if, Y - It u V - i a' - n' a i i i i D 1 1 1 1 1 / 0 O27»C»»2l»c * *I6 C D D oc 1 1 1 1 1 1 1 1 1 10 12 WEEKS Fig. 1. Weekly cumulative emergence of C. /. felis adults at various constant temperatures and 75% RH. within 20 days. In saturated air, 92% of the preemerged adults versus 62% of the emerged adults survived for 70 days. The humidity inside the cocoon did not differ from ambient; therefore, the cocoon did not provide a barrier to water loss. However, adult fleas inside cocoons were significantly less metabolically active than recently emerged adults, as indicated by CO 2 production (F < 0.0005; ANOVA, analysis of variance F (26) = 38.3). After 8 h, undisturbed pre-emerged adults produced an average of 0.15 ng CO 2 perflea.undisturbed and agitated emerged adults produced 2.9- and 3.5-fold more CO 2 than did pre-emerged adults, respectively. Pupae removed from cocoons or with the cephalic end of the cocoon dissected completed development and emerged in an average of 3.8 (n = 28) and 4.2 (n = 39) days, respectively (differences not significant, P = 0.05; Kruskal-Wallis Test, H = 36.8). It was observed that pre-emerged adults became active once the pupal cuticle was shed. When the pygidial areas of pupae were exposed through the cocoon, the average emergence time was significantly increased to 8.4 days (P < 0.05, n = 26), but it was not significantly different from the emergence time for fleas in intact cocoons (i =9.4 days; P > 0.05, n = 28). The cocoon, therefore, slightly impeded emergence. When pressure was applied to the outside of several cocoons, those adults not emerging immediately were observed moving within their cocoons, apparently attempting to emerge. Direct pressure ruptured the cocoon, facilitating adult exit. Experiment 3. With continuous exposure to 32 C, 60% of the adults emerged in the first week and all emerged within 4 weeks (Fig. 1). The rate of emergence at 27 and 21 C was similar to that at 32 C, with 80% emergence in 4 weeks. Emergence was considerably prolonged at 11 C; all adults emerged by week 20. The average time for emergence of adults from cocoons continuously a i
766 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 78, no. 6 Table 1. Effect of larval nourishment and number of cocoons spun on the weights of prepupae and the rate of adult emergence Table 2. Combined effect of a warm substrate and 38 g/cm 2 force applied on the emergence of the pre-emerged adult in the cocoon Condition of larvae Fully nourished Incompletely nourished Spun 1 cocoon Spun 2 cocoons f Prepupal wt (ngf Wet Dry 725 110 348 58 662 180 605 168 Adult emergence (days)" X + SD Median 58.8 ± 12.66 14.6 ± 15.76 57.2 ± 9.84 26.6 ± 6.37 56 8 70 8 " There were significant wet and dry wt differences between fully and incompletely nourished larvae (P < 0.01; t(i2) = 8.6) and between larvae that spun 1 versus 2 cocoons (P < 0.05; t(7) = 3.3) with the t test. ^ There were significant differences between fully and incompletely nourished larvae (C/(i5,i5) = 185.5; P < 0.001) and between larvae that spun one versus two cocoons (U 15,15) = 192; P < 0.001) with the Mann-Whitney test. exposed to 95, 75, and 33% RH took 18.2, 15.6, and 16.2 days, respectively (not significant, P = 0.05; ANOVA). Immature fleas reared at 91, 71, or 52% RH produced prepupae containing 77.0, 73.0, and 74.7% body water, respectively. Pre-emerged adults reared from larvae at the highest RH emerged significantly faster (x = 10.3 days) than did those reared at 71% RH (33.9 days) or 52% RH (25.5 days) (P < 0.05, ANOVA). Experiment 4. Prepupae that developed from incompletely nourished larvae were significantly lighter and contained less water than did prepupae from fully nourished larvae (P < 0.01, t test). Adults that developed from partially starved larvae emerged sooner than adults that developed from nonstressed larvae (P < 0.001, Mann-Whitney test, U (1515) = 185.5) (Table 1). Fifty percent of the adults produced from incompletely nourished larvae emerged in <8 days. However, several adults remained within their cocoons for up to 50 days. 13 26 38 52 64 96 127 191 FORCE APPLIED (9/cm 1 ) Fig. 2. Direct linear relationship between the percentage of C. /. felis adults emerging from cocoons and various single forces applied to the cocoon for 1 s (P < 0.0005, t = 6.08). 254 Substrate temp ( C) 38 35 32 29 25 i ± SD % adult emergence" After fo ce 87 ± 11.5a 57 + 12.2b 23 ± 17.3c 0d Od 93 ± 5.8a 97 + 5.8a 60 ± 36.0b 20 ± 10.0c 7 ± 5.8c 0 Means followed by the same letter within columns are not significantly different (P < 0.05; Duncan's [1955] multiple range test). Larvae forced to construct two cocoons developed into lighter prepupae and adults that emerged significantly sooner than individuals that spun only a single cocoon (Table 1). Experiment 5. No emergence was observed when the substrate containing pre-emerged adult fleas was vigorously vibrated. The vibrations created were much stronger than would normally be experienced by fleas in nature. A more common situation such as a human walking within 10 cm of a pre-emerged adult on the ground also failed to stimulate emergence. Experiment 6. There was a significant positive linear relationship between adult emergence and the amount of force applied to the cocoon (P < 0.005, t test, t a) = 6.08) (Fig. 2). There was no difference in rate of emergence when the force was applied for 1 or 30 s. Forces >254 g/cm 2 crushed most of the adults within the cocoons. Adults that did not emerge after a single force was applied moved within the cocoon for a few seconds before becoming quiescent. When the force was reapplied at 1-min intervals, more adults emerged. Adult emergence increased to 20% with a second application of 13 g/cm 2 force, but did not increase with additional applications of force. The slope of the linear regression was not significantly different from b = 0 at this force (t 3 = 1.73, P < 0.05) but was significantly lower than the other forces (P < 0.05). Adult emergence increased to 60% after 26 g/cm 2 was applied five times, whereas five applications of 38 g/cm 2 force resulted in 100% emergence. However, regression lines developed for the 38 and 26 g/cm 2 forces were not significantly different from each other (* 3 = 1.85, P < 0.05). A 75-kg human walking on the flea-infested carpeting initially stimulated 31% of the adult fleas to emerge. Successive force at 1-min intervals caused 63, 80, 97, and 100% cumulative adult emergence. Experiment 7. Most adults (87%) emerged within 5 s when cocoons were placed on a glass surface heated to 38 C. After a 38 g/cm 2 force was administered, an additional 6% of the adults emerged. Substrate temperatures of 35 and 32 C stimulated 57 and 23% emergence, respectively,
November 1985 SlLVERMAN AND RUST: LONGEVITY IN CAT FLEA 767 and 97 and 60% emergence, respectively, after the force was applied. Lower surface temperatures (29 C) and ambient (25 C) did not stimulate emergence and few adults emerged after the force was applied (Table 2). Discussion The pre-emerged adult stage of C. /. felis enclosed within the cocoon, yet capable of rapid emergence, is an ideal adaptation to increase the likelihood that the flea will survive host-free periods and successfully encounter a mobile host. Prolonged adult survival within the cocoon, particularly under desiccating conditions, is due to quiescent periods of low metabolic activity rather than restriction of water loss through the cocoon wall. Bursell (1974) stated that conditions of activity that increase oxygen demand increase spiracular opening to meet respiratory requirements, with a resultant loss of water from the tracheal system. It is likely that, since CO 2 production was lower in pre-emerged adults, the spiracles opened less frequently and loss of water through the spiracles was reduced. Pupae removed from their cocoons yield active adults immediately following the pupal-imaginal molt. The cocoon impedes adult emergence somewhat and may also protect adult fleas from nonhost-produced stimuli. Sgonina (1939) determined that air currents received by sensory hairs on the pygidium stimulated movement of the emerged adult European hedgehog flea, Archaeopsylla erinacei (Bouche). We found that emerged adult C. /. felis jumped in response to a short burst of air, whereas adults within the cocoon were not stimulated sufficiently to cause emergence (unpublished data). Since air currents may not necessarily signal the presence of a host nearby, fleas jumping in response to air movement would expend metabolic reserves and probably reduce adult longevity. Since the cocoon deflects air movement, adult activity is not stimulated. Although there was a direct relationship between temperature and rate of adult C. /. felis emergence, at a given temperature there was also a proportion of the flea population that remained in the cocoon for extended periods. These observed differential rates of emergence of fleas held under similar environmental conditions are thought to result at least in part from larval competition for food. We have shown that adults that developed from light prepupae remained in their cocoons for a shorter time than adults from more robust prepupae. Two explanations are proposed: 1) emergence mechanisms are triggered when water and food reserves drop below a critical level, or 2) an incompletely nourished prepupa may produce a weaker cocoon that does not impede adult emergence. The extended survival of pre-emerged adults within cocoons is of little consequence if a host is not eventually located. We believe the primary factor responsible for initiating adult emergence and reducing the randomness of host location is host-produced stimuli. Margalit and Shulov (1972) observed that adult rat fleas emerged earlier when stimulated even though the source of stimulation was not described. Similarly, the activity of the sand martin is responsible for the emergence of the overwintering bird flea Ceratophyllus styx (Humphries 1969). Frequent hosts of cat fleas such as domestic and feral cats and dogs, mustelids, and opossums do not necessarily return to flea-infested lairs. Thus, successful attack of a mobile host necessitates immediate emergence and host-seeking behavior. Contrary to Humphries' (1969) observation with sand martin fleas, no emergence of C. /. felis was observed when substrates containing pre-emerged fleas were vigorously vibrated; however, direct pressure on the cocoon stimulated emergence. Since the combination of warmth and pressure provided higher emergence rates than either warmth or pressure alone, it is likely that an endothermic animal resting on a cocoon increases the chance that the adult flea would emerge and successfully attack a host. Acknowledgment We thank Rudolf H. Scheffrahn, Arthur G. Appel, and Donald A. Reierson for their advice and critical reviews of the manuscript. We are also indebted to the Statewide Critical Applied Research funds for partial support of this study. References Cited Bacot, A. W. 1914. A study of the bionomics of the common rat fleas and other species associated with human habitations, with special reference to the influence of temperature and humidity of various periods in the life cycle of the insect. J. Hyg. 13 (Plague Suppl. 3): 447-654. Bursell, E. 1974. Environmental aspects humidity, pp. 44-84. In M. Rockstein [ed.], The physiology of insects, vol. 2. Academic, New York. Duncan, D. B. 1955. Multiple range and multiple F tests. Biometrics 11: 1-41. Hopkins, G. H. E., and M. Rothschild. 1953. An illustrated catalogue of the Rothschild collection of fleas in the British Museum, vol. 1. University, Cambridge. Humphries, D. A. 1969. Behavioral aspects of the ecology of the sand martin flea Ceratophyllus styx jordani Smit (Siphonaptera). Parasitology 59: 311-344. Karandikar, K. R., and D. M. Munshi. 1950. Lifehistory and bionomics of the catflea,ctenocephalides felis Bouche. J. Bombay Nat. Hist. Soc. 49: 169-177. Margalit, J., and A. S. Shulov. 1972. Effect of temperature on the development of prepupa and pupa of the rat flea, Xenopsylla cheopis Roths. J. Med. Entomol. 9: 117-125. Sgonina, K. 1939. Wirtsfindung and wirtsspezifitat
768 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 78, no. 6 von flohen, pp. 1663-1668. In Transactions, Inter- Winston, P. W., and D. H. Bates. 1960. Saturated national Congress of Entomology, Berlin, 1938. solutions for the control of humidity in biological Silverman, J., M. K. Rust, and D. A. Reierson. 1981. research. Ecology 41: 232-237. Influence of temperature and humidity on survival and development of the cat flea, Ctenocephalides Received for publication 14 January 1985; accepted felis (Siphonaptera: Pulicidae). J. Med. Entomol. 18: 22 May 1985. 78-83.